Capnometers can fail in a number of ways that are not detectable using current static calibration testing procedures. These failures could include leaks in the sample line, a weak sample pump, inadequate dynamic response, flow restriction and flow obstruction. Failure of a capnometer during critical care procedures such as sedation and anesthesia can be life threatening to patients. We plan to develop a patient breath simulator that will allow for the testing of capnometers under dynamic conditions seen during patient use. We propose to develop a portable device and control algorithms that will be capable of simulating CO2 concentrations and pressure waveforms mimicking human respiratory cycles. The same device will have additional testing modes that will identify capnometer failures that cannot be detected under static calibration. We propose to verify the ability of the device to detect real world failure modes by testing it on a number of capnometers that have been modified to simulate these failures. The ability to detect capnometer failures under real world conditions using an automated tester will improve patient safety as well as increase standardization of testing procedures and decrease testing and calibration costs.